Pharmacological Strategies for Improving Diastolic Dysfunction in the Setting of Chronic Pulmonary Hypertension
Background—Right ventricular (RV) hypertrophy is an adaptive process that occurs in the setting of chronic pulmonary hypertension (CPH) and can lead to alterations in normal RV diastolic properties. This study was designed to investigate the effects of NO and milrinone on RV diastolic dysfunction in the setting of CPH and RV hypertrophy by use of a canine model of monocrotaline pyrrole (MCTP)–induced CPH.
Methods and Results—Sixteen mongrel dogs (22 to 24 kg) were used. Animals underwent percutaneous pulmonary artery (PA) catheterization to measure pulmonary hemodynamics before and 8 weeks after injection of 3 mg/kg MCTP (n=8) or placebo (control, n=8). Eight weeks after injection, all hearts were instrumented with a PA flow probe, sonomicrometric dimension transducers, and micromanometers. Data were collected at baseline and after both NO and milrinone administration. Diastolic properties were quantified by use of the end-diastolic pressure-volume relationship and the time constant of ventricular isovolumic relaxation. Eight weeks after injection, significant increases in the PA pressure and pulmonary vascular resistance were observed in MCTP dogs. Significant worsening of RV diastolic function occurred in association with significant increases in the ratio of RV dry weight to LV+septal dry weight. NO and milrinone administration both led to significant improvements in RV diastolic properties.
Conclusions—In the setting of MCTP-induced CPH, significant worsening of RV diastolic function was observed in association with significant increases in the ratio of RV dry weight to LV+septal dry weight, suggesting that these changes are partially due to RV hypertrophy. The significant improvement in RV diastolic properties after both NO and milrinone administration suggests that these agents may be effective forms of pharmacological therapy for improving RV diastolic dysfunction in the setting of CPH.
The clinical management of diastolic dysfunction is challenging, and despite the existence of several therapeutic options, no single treatment is universally recommended. Contemporary clinical strategies for improving abnormal ventricular diastolic properties include the use of pharmacological agents that increase ventricular relaxation and ease cardiac loading conditions.1 2 These strategies, however, have been applied primarily toward treating LV diastolic dysfunction in the setting of systemic arterial hypertension and myocardial hypertrophy. Few reports describe the application of these measures with respect to improving RV diastolic dysfunction associated with cardiac hypertrophy in the setting of CPH.
Nitric oxide is not only a selective pulmonary vasodilator without significant hemodynamic effect on the systemic circulation3 4 but also an important mediator of cardiac diastolic function.5 6 Milrinone, conversely, is a phosphodiesterase inhibitor that possesses positive inotropic and selective pulmonary vasodilatory effects.7 In addition to these properties, milrinone is also a positive lusitropic agent.8 9 Such properties suggest that both these agents may be potentially useful strategies for improving RV diastolic dysfunction in the setting of chronically elevated pulmonary vascular pressures. However, an appropriate large-animal model of CPH and RV hypertrophy has not previously been available for basic investigation of this problem. This study was therefore designed to investigate the effects of NO and milrinone on RV diastolic dysfunction in the setting of CPH and RV hypertrophy by use of a canine model of MCTP-induced CPH10 and functional assessment of RV chamber stiffness.
Hemodynamic Monitoring, Experimental Groups, and Drug Synthesis and Injection
The anesthetic regimens, mode of ventilatory support, and all invasive hemodynamic and metabolic monitoring used in this investigation have previously been well described.10 Hemodynamic measurements were done at baseline and 8 weeks after monocrotaline injection. MCTP or placebo was injected into the right atrium.
Experimental animals (MCTP, n=8, 22 to 24 kg) received an injection of 3 mg/kg MCTP dissolved in N,N-dimethyl formamide at a concentration of 40 mg/mL, and control animals (n=8, 22 to 24 kg) received dimethyl formamide as placebo. Consequently, the volume of injectate was dependent on the individual weight of each animal and varied from 1.65 to 1.8 mL. MCTP was synthesized by a previously well-established method.11
Instrumentation and Surgery
Eight weeks after MCTP injection, a standard median sternotomy and an anterior pericardiotomy were performed to expose the hearts of all animals. An ultrasonic flow probe (T208X, Transonic Systems Inc) was placed around the main pulmonary trunk to measure pulmonary blood flow. Micromanometers (MPC-500, Millar Instruments Inc) were placed into the RV and LV, right and left atrium, and pulmonary artery for continuous recording of intracavitary pressures.
Hemispheric ultrasonic dimension transducers (1.5-mm OD, No. 1–1015-5A, Vernitron) were sewn to the epicardial surface of the heart across the base-apex major-axis and anteroposterior minor-axis diameters of the LV. Two additional transducers were also placed on the epicardial surfaces of the RV and LV free walls, and another was inserted into the interventricular septum to measure the septal–free wall minor-axis diameters of both the RV and LV.
Experimental Protocol During Pharmacological Studies
Eight weeks after MCTP injection, baseline functional and hemodynamic data were collected in every animal in the control and MCTP groups after the instrumentation procedure. Control animals did not receive any pharmacological treatment or undergo any further data collection. In MCTP animals, inhaled NO (NO, 777 ppm and NO2, <0.1 ppm, National Specialty Gases) was then administered into the ventilator at levels of 40 and 80 ppm, and data were again collected at each respective concentration. NO and NO2 levels were measured by continuous chemiluminescent analysis (model 42H, Thermo Environmental Instruments, Inc).
After each incremental change in the NO concentration, the animal’s condition and hemodynamic parameters were allowed to equilibrate for 10 minutes before any further data collection. After data were obtained at the dose of 80 ppm, NO was stopped, and the level in the ventilator’s inspiratory circuit was restored to 0 ppm. Pulmonary hemodynamic parameters were allowed to return to baseline, which usually occurred within 5 to 8 minutes of discontinuation.
Milrinone was then loaded as an initial dose of 50 μg · kg−1 · min−1 given over 10 minutes and subsequently infused at rates of 0.5 μg · kg−1 · min−1 and 1.0 μg · kg−1 · min−1. As with NO, after each incremental increase in the milrinone infusion rate, pulmonary hemodynamic parameters were permitted to stabilize for 10 minutes before additional data acquisition. The entire duration of anesthesia, including instrumentation, drug administration, and all data acquisition, was ≈2.5 to 3 hours in each MCTP animal.
Raw data were digitized on-line, collected, and stored on a microprocessor (PDP 11/23, Digital Equipment Corp). All data were analyzed on a Dell Dimension XPS P90 computer (Dell Computer Corp) with software that has previously been well described.12
After collection of hemodynamic and functional data, all hearts were excised, and the atria were separated from the ventricles at the atrioventricular groove. The RV free wall was separated from the LV and septum. The RV and LV from each animal were weighed, and wall volumes were then determined by saline displacement. To calculate percent water, all samples were placed in a 120°C oven and dried for 24 hours. The dry weight of each tissue sample, for both the RV and LV+septum, was measured, and these weight ratios were compared between experimental groups. The percentage of tissue water content was calculated from the following equation: (wet weight−dry weight×100%)/wet weight.
Calculation of Ventricular Cavitary Volumes
LV and RV cavitary volumes were calculated by use of well-validated models.12 13 LV cavitary volume was calculated by representing the epicardial surface of the LV as an ellipsoidal shell and subtracting ventricular wall volume according to the following equation: LVV=πb2a/6−Vwall, where b is the minor-axis diameter, a is the major-axis diameter, and Vwall is LV wall volume.12 RV cavitary volume was calculated according to the ellipsoidal shell subtraction method,13 which assumes an ellipsoidal shape for the LV as well as the total biventricular shell, by the following equation: RVVendo=BVVepi−LVVepi−FWV, where BVVepi is the biventricular epicardial shell volume, LVVepi is the LV epicardial shell volume, and FWV is the volume of the RV free wall.13
Assessment of Biventricular Diastolic Properties
Chamber stiffness was quantified by fitting the exponential function EDP=A×eαEDV to EDP and EDV obtained during vena caval occlusion for both the RV and LV by use of nonlinear least-squares regression. α is defined as the change in the logarithmic function of ventricular pressure relative to the change in volume and represents an index of ventricular chamber stiffness, whereas A is a proportionality constant.
A parameter of myocardial stiffness was quantified according to the following equation: myocardial stiffness=αVw, where α is chamber stiffness and Vw is ventricular wall mass.14
To assess ventricular isovolumic pressure decay, the period of isovolumic relaxation was defined from the nadir of the ventricular pressure derivative (dP/dtmin) to the point at which ventricular pressure decreased to the level of atrial pressure.15 The time constant of isovolumic pressure decay (τ) was then calculated from the following monoexponential model: PIVR=P0e−t/τ+P∞, where P0 is the initial pressure above P∞ during isovolumic relaxation, t is time, P∞ is the asymptote to which pressure decays, and τ is the time required for pressure at any point on the isovolumic curve to fall to 1/e of its original value.14 RV/LV model parameters were estimated by nonlinear least-squares regression.
Assessment of Biventricular Systolic Function
RV and LV systolic function were assessed with load-insensitive means. The highly linear relationship between stroke work and end-diastolic chamber volume was calculated from data acquired during vena cava occlusion by use of least-squares linear regression. The slope of these linear regressions is known as the PRSW and represents a load-independent index of systolic function and myocardial contractility.12
Experimental Approval and Animal Rights
The experimental setup and procedures conformed to the guidelines established by the American Physiological Society and the National Institutes of Health (Guide for the Care and Use of Laboratory Animals, National Institutes of Health publication 86–23, revised 1985). The experiments described in this report were approved by the Duke University Institutional Animal Care and Use Committee (DUIACUC Assigned Registry A621–95-9R1).
Statistical analysis was performed with commercially available software (SigmaStat Version 2.0, Jandel Corp). Hemodynamic data obtained before and after injection of MCTP within the MCTP group were analyzed with standard two-tailed paired Student’s t tests. Unpaired Student’s t tests were used to compare all data between MCTP and control.
To test for significant changes in the parameters of diastolic function as well as systolic function after NO inhalation within the MCTP group, a one-way ANOVA for repeated measurements was used. Follow-up comparison of each index of ventricular diastolic properties at the individual levels of NO were done with paired Student’s t tests. Comparison of changes in RV diastolic function occurring after milrinone infusion was performed in a similar fashion. The results are expressed as mean±SEM. A difference was considered statistically significant at P<.05.
There was no significant difference in the baseline pulmonary hemodynamic indices between control and MCTP dogs; however, 8 weeks after MCTP injection, significant differences in these same indices were observed between the two groups (Table 1⇓).
Diastolic Properties in MCTP-Induced CPH
The exponential EDP-EDV model accurately represented (r2>.92) all pressure and volume data during transient vena cava occlusion in both groups. No significant differences in LV α were observed between control (.021±.005) and MCTP (.024±.005). There was, however, a significant increase in the RV α of MCTP (.070±.006, P<.05) compared with control (.046±.008). RV αVw was significantly greater in MCTP (.724±.066) than in control (.383±.068, P<.005), while no significant difference was observed in LV αVw between the two groups (.560±.108 versus .495±.105).
The monoexponential model of isovolumic pressure decay accurately predicted (r2>.993) the fall of ventricular pressure during isovolumic relaxation for the RV and LV. There was no significant difference in LV τ (41±3 versus 42±3 ms) between the two experimental groups. RV τ was significantly longer, however, in MCTP (84±15 ms, P<.05) than in control (35±3 ms).
Systolic Function in MCTP-Induced CPH
Highly linear relationships (r2>.95) were obtained between calculated RV and LV EDV and stroke work during vena caval occlusion in both control and MCTP 8 weeks after MCTP injection. There was no significant difference in LV PRSW between MCTP (94.65±6.26×103 erg/mL) and control (87.98±7.23×103 erg/mL). However, significant increases in RV PRSW were observed in MCTP (46.90±3.16×103 erg/mL, P<.0005) compared with the control values (24.56±3.12×103 erg/mL).
Effect of NO on Diastolic Properties and Systolic Function
NO led to significant improvements in pulmonary hemodynamic indices (Table 2⇓). In addition, significant decreases occurred in RV α as well as RV αVw, whereas no significant improvements in these same parameters were observed in the LV (Table 3⇓). RV and LV τ were significantly shorter after NO inhalation; however, RV/LV PRSW was not significantly different (Table 3⇓). Although NO did not lead to significant changes in the LV α of MCTP dogs, a substantial downward shift was observed in the LV EDP-EDV relationship of these animals (Fig 1⇓). The changes occurring in the EDP-EDV relationship as a result of NO inhalation for both the LV and RV are illustrated in Figs 1⇓ and 2⇓, respectively.
Effect of Milrinone on Diastolic Properties and Systolic Function
Milrinone infusion caused significant decreases in the pulmonary vascular resistance, which were associated with significant improvements in pulmonary blood flow (Table 4⇓). Although there were decreases in the mean pulmonary artery pressure at 0.5 and 1.0 μg · kg−1 · min−1, these differences were not statistically significant.
Milrinone infusion also led to significant improvements in RV α, RV αVw, RV/LV τ, and RV/LV PRSW (Table 5⇓). There was no significant change in LV α after milrinone infusion; however, as with NO, a substantial downshifting of the EDP-EDV relationship was also noted (Fig 3⇓). The changes occurring in the EDP-EDV relationship after milrinone infusion for both the LV and RV are illustrated in Figs 3⇓ and 4⇓, respectively.
Tissue Analysis in the Setting of MCTP-Induced CPH
There was a significant increase in the RV dry weight of MCTP dogs (10.29±0.55 g, P<.05) compared with control animals (8.40±0.51 g). Both the ratio of RV dry weight to LV+septal dry weight (0.45±0.02 versus 0.37±0.01, P<.05) and the ratio of RV dry weight to body weight (0.43±0.02 versus 0.36±0.02 g RV/kg body wt, P<.05) were also significantly elevated in MCTP animals compared with the control group. No significant differences were observed in the LV dry weights (23.3±1.4 versus 23.3±1.3 g) or in the RV (79.6±0.4% versus 78.6±0.4%) and LV (79.0±0.4% versus 78.4±0.3%) water contents between the two experimental groups.
CPH represents an important clinical sequela that commonly occurs as a result of chronic obstructive pulmonary disease. It is also associated with a number of other pathological processes, including long-standing congestive heart failure, mitral valve disease, congenital heart defects, and chronic pulmonary embolism; alternatively, CPH may exist as a primary process. This state of chronically elevated pulmonary vascular pressures can lead to adaptive mechanisms in the RV and eventually result in altered systolic and diastolic mechanics.
Myocardial hypertrophy is a compensatory process of the heart that can develop in the setting of a sustained pressure overload, such as CPH, and is perhaps one of the most important causes of diastolic dysfunction.2 Hypertrophied myocardium is particularly susceptible to diastolic dysfunction as a result of structural changes (increased LV mass) and impaired ventricular relaxation.16 In a previous investigation, a significant correlation was noted between myocardial hypertrophy and diastolic dysfunction, which later improved after regression of the hypertrophy.17
Approximately 10 years ago, NO was identified as an important endogenous biological mediator of vascular tone.18 19 Since that initial discovery, clinical3 and experimental4 investigations have shown that NO is a selective pulmonary vasodilator without significant effect on the systemic circulation. Recent reports have also suggested that NO is an important regulator of cardiac diastolic function.5 6
Milrinone is a phosphodiesterase inhibitor that possesses positive inotropic, selective pulmonary vasodilator, and positive lusitropic effects.8 In a canine model of embolism-induced pulmonary hypertension using autologous muscle, milrinone increased cardiac index while improving mean pulmonary artery pressure and pulmonary vascular resistance.20 Clinically, in the setting of congestive heart failure, significant improvements have been observed in the indexes of LV diastolic performance as well as overall global improvement of diastolic function after administration of milrinone.9 21
Thus, NO and milrinone may both be potentially useful pharmacological strategies for improving altered RV diastolic properties in the setting of CPH. However, experience with these two agents in treating RV diastolic dysfunction is currently limited. Experimental investigation of this problem has not previously been possible because of lack of an appropriate large-animal model of CPH.
In this investigation, the effects of NO and milrinone on RV diastolic and systolic mechanics were assessed in a recently established canine model of MCTP-induced CPH.10 MCTP-induced CPH is characterized by a proliferative vasculitis. Pulmonary vascular injury, after the initial injection, affects the medium and small pulmonary arterioles. Subsequent vascular remodeling leads to endothelial degeneration and hyperplasia, smooth muscle hypertrophy and medial thickening, and perivascular connective tissue proliferation.22 The development of these microscopic changes along with the ensuing rise in pulmonary vascular pressures is gradual. In addition, the characteristic parenchymal features of this model are often observed in the pulmonary vasculature of patients with pulmonary hypertension of differing etiologies, including end-stage cardiac disease as well as a subtype of primary pulmonary hypertension known as plexogenic pulmonary arteriopathy.23
MCTP-induced CPH has a significant impact on RV diastolic properties and an insignificant effect on LV diastolic function. RV α and RV αVw were significantly elevated in animals with MCTP-induced CPH, whereas no significant differences were observed in LV α or αVw. The significant increase in RV α should be interpreted with caution, because it may have been due to collagen recruitment. It is unlikely, however, that these changes were the result of MCTP-induced myocardial damage, because previous morphological analysis using light and electron microscopy suggests that monocrotaline itself has no direct toxic effect on myocardial tissue.24
Ventricular chamber stiffness, as estimated by the EDP-EDV relationship, represents a useful means for quantifying diastolic function in an intact heart.25 Pressure-volume relationships provide important information regarding diastolic function26 and have been applied in clinical and experimental investigations.27 28 Another technique that is often used for assessing diastolic properties involves assessment of myocardial elasticity with parameters of stress and strain.25 In a previous investigation comparing patients with normal and hypertrophied hearts, however, myocardial diastolic stiffness, as defined by use of stress-strain relations, remained relatively unchanged despite dramatic shifts in the passive pressure-volume relations.29 Measurement of pressure-volume relations may therefore be a more sensitive means to assess myocardial diastolic properties than estimation of stress-strain relations.30
τ represents another useful index frequently used for assessing diastolic properties that has been shown to be independent of any changes in loading conditions15 31 and is relatively insensitive to heart rate.32 33 34 RV τ was significantly longer in the setting of MCTP-induced CPH. However, no significant differences were observed in LV τ between the two experimental groups.
Tissue dry weight analysis revealed that the RV was significantly hypertrophied, whereas no significant change in LV muscle mass occurred. There were significant elevations in the ratio of RV dry weight to LV+septal dry weight as well as the ratio of RV dry weight to body weight in MCTP animals compared with control animals, indicating that a significant degree of RV hypertrophy had occurred. These findings suggest that the alterations in RV diastolic and systolic properties observed in this model are at least partially due to hypertrophy. The significant augmentation of RV systolic function associated with the significant worsening of RV diastolic properties may provide insights into the mechanisms of myocardial dysfunction in the setting of CPH.
NO-mediated improvement of ventricular diastolic properties can occur indirectly through easing of RV loading conditions or via direct myocardial relaxation. It has previously been shown that endogenous NO released from coronary endothelial cells plays an important role in the modulation of myocardial relaxation activity.35 In other investigations using sodium nitroprusside as an exogenous donor, NO was shown to exert a direct myocardial relaxant effect in isolated ejecting hearts, independent of any changes in cardiac loading,36 as well as in isolated papillary muscles.37 Similar results were observed in the clinical setting of Paulus and associates,5 where significantly improved LV diastolic properties were observed after global intracoronary infusion of sodium nitroprusside.
In this report, however, the beneficial effects of NO appear to be related to its well-established vasodilator activity. By improving pulmonary hemodynamic indices, NO led to a significant easing of RV loading conditions, which were observed in association with significant decreases in RV diastolic properties and LV τ. Although the use of sodium nitroprusside as an exogenous NO donor might be equally effective in decreasing RV α and RV τ in this model, such conclusions are not possible on the basis of the data presented in this study. Future studies should examine whether sodium nitroprusside would be as effective as inhaled NO for improving RV diastolic properties in MCTP-induced CPH.
Although the lusitropic effects of milrinone have been clearly demonstrated in this study in association with significant improvements in pulmonary blood flow, the vasodilatory properties of this agent can potentially play a key role in its total effects and may partially account for the observed improvements in RV α, RV τ, and overall hemodynamic status in this model. In a previous report by Monrad and associates,9 the relative ratio of the lusitropic and vasodilator actions of milrinone was examined in patients with advanced congestive heart failure. In that investigation, significant improvement of LV diastolic function was observed in association with relatively small decreases in mean arterial pressure, suggesting that the influence of milrinone on cardiac relaxation at the level of the individual myocyte plays a more fundamental role in its overall effects than its vasodilator actions. In the present report, significant improvements in RV diastolic properties as well as LV τ occurred in association with slight decreases in mean pulmonary artery pressure after milrinone infusion, which would appear to be consistent with the findings of Monrad and associates. Further investigation should attempt to elucidate the relative contributions of the lusitropic and vasodilatory effects of milrinone in MCTP-induced CPH.
Evaluation of diastolic function with in vivo models necessitates careful measurement of the dynamic changes occurring in ventricular pressure and volume within an intact heart. With respect to EDP, use of transmural pressures is required for an accurate assessment of diastolic properties.31 Estimates of α from EDP-EDV relationships, however, generally ignore any possible restrictive effect of the pericardium and assume that no external forces are imposed on the heart when intraventricular pressure is elevated.38 When the pericardium is left intact, the possibility exists that pericardial pressures might exceed zero and violate these assumptions. Previous investigations have demonstrated that in the setting of pressure or volume overload, an intact pericardium contributes significantly to increases in EDP and may affect the EDP-EDV relationship independently of changes in ventricular diastolic properties.39 40 Admittedly, use of an open-chest, open-pericardium model could potentially introduce a degree of experimental artifact and might be considered a study limitation. It should be noted, however, that this type of experimental preparation allows one to obtain true transmural pressures, which, as previously discussed, are requisite for accurate assessment of ventricular diastolic properties.
It is acknowledged that the absolute level of pulmonary hypertension occurring after MCTP injection may not accurately reflect the degree of CPH observed clinically. Consequently, the amount of hypertrophy induced in this model was much less than what one would expect in the clinical setting. It is important to point out, however, that the baseline pulmonary vascular pressures observed in the dogs before drug injection were significantly lower than normal levels found in human beings and that a twofold increase in pulmonary hemodynamic indices was achieved. Even at this level of pulmonary hypertension, significant alterations in RV diastolic properties were observed, which improved significantly after pharmacological therapy. A potential way to achieve a higher degree of pulmonary hypertension would be to perform multiple injections of MCTP at regular intervals to theoretically cause repeated injury to the pulmonary vascular bed and produce greater scarring of the pulmonary circulation. Higher pulmonary hemodynamic indices might, in turn, lead to more severe hypertrophy. When such a model of CPH is used for basic investigation, however, it is reasonable to state that the resultant pulmonary hypertension should not be so severe that clinical right-sided heart failure would ensue and result in a highly unstable situation.
In summary, a canine model of MCTP-induced CPH provides a useful means to evaluate various pharmacological strategies for improving RV diastolic dysfunction in the setting of CPH. There was a significant worsening of RV diastolic properties, which was significantly improved after NO and milrinone administration. These data suggest that NO and milrinone may both be effective pharmacological strategies in the treatment of diastolic dysfunction in the setting of CPH.
Selected Abbreviations and Acronyms
|CPH||=||chronic pulmonary hypertension|
|LV||=||left ventricular, left ventricle|
|PRSW||=||preload recruitable stroke work|
|RV||=||right ventricular, right ventricle|
Dr Chen is a recipient of a National Research Service Award, fellowship HL-09489.
Reprint requests to Edward P. Chen, MD, Department of Surgery, S-343, University of California San Francisco, 505 Parnassus Ave, San Francisco, CA 94143.
Presented at the 69th Scientific Sessions of the American Heart Association, New Orleans, La, November 10–13, 1996, and published in abstract form (Circulation. 1997;96(suppl I):I-63).
- Received September 18, 1997.
- Revision received October 21, 1997.
- Accepted November 6, 1997.
- Copyright © 1998 by American Heart Association
Frostell C, Fratacci M-D, Wain JC, Jones R, Zapol WM. Inhaled nitric oxide: a selective pulmonary vasodilator reversing hypoxic pulmonary vasoconstriction. Circulation. 1991;83:2038–2047.
Paulus WJ, Vantrimpont PJ, Shah AM. Acute effects of nitric oxide on left ventricular relaxation and diastolic distensibility in humans. Circulation. 1994;89:2070–2078.
Monrad ES, McKay RG, Baim DS, Colucci WS, Fifer MA, Heller GV, Royal HD, Grossman W. Improvement in indexes of diastolic performance in patients with congestive heart failure treated with milrinone. Circulation. 1984;70:1030–1037.
Glower DD, Spratt JA, Snow ND, Kabas JS, Davis JW, Olsen CO, Tyson GS, Sabiston DC, Rankin JS. Linearity of the Frank-Starling relationship in the intact heart: the concept of preload recruitable stroke work. Circulation. 1985;71:994–1009.
Feneley MP, Elbeery JR, Gaynor JW, Gall SA, Davis JW, Rankin JS. Ellipsoidal shell subtraction model of right ventricular volume. Circ Res. 1990;67:1427–1436.
Frederiksen JW, Weiss JL, Weisfeldt ML. Time constant of isovolumic pressure fall: determinants in the working left ventricle. Am J Physiol. 1978;235:H701–H706.
Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A. 1987;84:9265–9269.
Edwards WD, Edwards JE. Clinical primary pulmonary hypertension: three pathologic types. Circulation. 1977;56:884–888.
Glantz SA, Parmley WW. Factors which affect the diastolic pressure-volume curve. Circ Res. 1978;42:171–180.
Parmley WW, ed. Cardiology. Philadelphia, Pa: JB Lippincott Co; 1992:chap 5.
Grossman W, McLaurin LP, Stefadouros MA. Left ventricular stiffness associated with chronic pressure and volume overloads in man. Circ Res. 1974;35:793–800.
Mathey D, Bleifeld W, Franken G. Left ventricular relaxation and diastolic stiffness in experimental myocardial infarction. Cardiovasc Res. 1974;8:583–592.
Parmley WW, Spann JF, Taylor RR, Sonnenblick EH. The stress elasticity of cardiac muscle in hyperthyroidism, ventricular hypertrophy, and heart failure. Proc Soc Exp Biol Med. 1968;127:606–609.
Parmley WW, ed. Cardiology. Philadelphia, Pa: JB Lippincott Co; 1992:chap 4.
Frais MA, Bergman DW, Kingma I, Smiseth OA, Smith ER, Tyberg JV. The dependence of the time constant of left ventricular isovolumic relaxation (τ) on pericardial pressure. Circulation. 1990;81:1071–1080.
Weiss JL, Frederiksen JW, Weisfeldt ML. Hemodynamic determinants of the time-course of fall in canine left ventricular pressure. J Clin Invest. 1976;58:751–760.
Weisfeldt ML, Frederiksen JW, Yin FCP, Weiss JL. Evidence of incomplete left ventricular relaxation in the dog: prediction from the time constant for isovolumic pressure fall. J Clin Invest. 1978;62:1296–1302.
Thompson DS, Waldron CB, Juul SM, Naqvi N, Swanton RH, Coltart DJ, Jenkins BS, Webb-Peploe MM. Analysis of left ventricular pressure during isovolumic relaxation in coronary artery disease. Circulation. 1982;65:690–697.
Grocott-Mason R, Anning P, Evans H, Lewis MJ, Shah AM. Modulation of left ventricular relaxation in isolated ejecting heart by endogenous nitric oxide. Am J Physiol. 1994;267:H1804–H1813.
Grocott-Mason R, Fort S, Lewis MJ, Shah AM. Myocardial relaxant effects of exogenous nitric oxide in isolated ejecting hearts. Am J Physiol. 1994;266:H1699–H1705.
Brodie BR, Chuck L, Klausner S, Grossman W, Parmley W. Effects of sodium nitroprusside and nitroglycerin on tension prolongation of cat papillary muscle during recovery from hypoxia. Circ Res. 1976;39:596–601.
Mirsky I, Rankin JS. The effects of geometry, elasticity, and external pressures on the diastolic pressure-volume and stiffness-stress relations: how important is the pericardium? Circ Res. 1979;44:601–611.
Spotnitz HM, Kaiser GA. The effect of the pericardium on pressure-volume relations in the canine left ventricle. J Surg Res. 1971;11:375–380.
Shirato K, Shabetai R, Bhargava V, Franklin D, Ross J. Alteration of the left ventricular diastolic pressure-segment length relation produced by the pericardium: effects of cardiac distension and afterload reduction in conscious dogs. Circulation. 1978;57:1191–1198.